Mohamed Hamada

A New Implementation for High Speed Normalized Neural Networks in Frequency Space

Neural networks have shown good results for detection of a certain pattern in a given image. In our previous work, a fast algorithm for object/face detection was presented. Such algorithm was designed based on cross correlation in the frequency domain between the input image and the weights of neural networks. In this paper, a simple design for solving the problem of local subimage normalization in the frequency domain is presented. This is done by normalizing the weights in the spatial domain off line. Furthermore, it is proved that local subimage normalization by normalizing the weights is faster than subimage normalization in the spatial domain. Moreover, the overall speed up ratio of the detection process is increased as the normalization of weights is done off line.

New Fast Decision Tree Classifier for Identifying Protein Coding Regions

In this paper, a fast tool for finding protein coding regions is presented. Such tool relies on performing cross correlation in the frequency domain and decision Tree. In addition, a modified trust region method is used to find the closet (optimized) DNA nucleotide. Moreover, a Sequential PRM-based protein folding algorithm for finding the point where these proteins add to the ladder is introduced. Furthermore, standard parallel scan algorithm is used to provide parallel processing of the strides and its transitions. This proposed tool produces more accurate results, than that have previously been obtained for a range of different sequence lengths. Experimental results confirm the scalability of the proposed classifying tool to handle large volume of datasets irrespective of the number of classes, tuples and attributes. High classification accuracy is achieved. The main achievement in this paper is the fast decision tree algorithm. Such algorithm relies on performing cross correlation in the frequency domain between the input data at each node and the input weights of neural networks. It is proved mathematically and practically that the number of computation steps required for the presented FNNs is less than that needed by conventional neural networks (CNNs). Simulation results using MATLAB confirm the theoretical computations.

Fast principal component analysis for face detection using cross-correlation and image decomposition

In a previous paper [24], fast PCA implementation for face detection based on cross-correlation in the frequency domain between the input image and eigenvectors was presented. Here, this approach is developed to reduce the computation steps required by fast PCA. The principle of divide and conquer strategy is applied through image decomposition. Each image is divided into small in size sub-images and then each one is tested separately by using a single fast PCA processor. In contrast to using only fast PCA, the speed up ratio is increased with the size of the input image when using fast PCA and image decomposition. Simulation results demonstrate that our proposal is faster than the conventional and Fast PCA. Moreover, experimental results for different images show good performance.

Fast Time Delay Neural Networks for Detecting DNA Coding Regions

In this paper, a new approach for fast information detection in DNA sequence has been presented. Our approach uses fast time delay neural networks (FTDNN). The operation of these networks relies on performing cross correlation in the frequency domain between the input data and the input weights of neural networks. It is proved mathematically and practically that the number of computation steps required for the presented FTDNNs is less than that needed by conventional time delay neural networks (CTDNNs). Simulation results using MATLAB confirm the theoretical computations.

Recommending Learning Peers for Collaborative Learning through Social Network Sites

With advances in social network sites and easy access to internet services, many learners depend on suggestions from other people on the internet for easy access to very essential information concerning learning materials, and also to collaborate with each other in order to exchange ideas. Current recommender systems for learning focus on recommending a sequence of learning materials based on learners similarities or similarities between the new learning objects and the ones the user is already familiar with in the past. Many learners prefer collaborative learning than learning on their own or in the classroom, but the major difficulty in engaging in an online collaborative learning is how to get a suitable collaborating partners(learning peers). This paper proposed a recommendation system that can search social network sites to find and recommend learning peers to the user based on their post, comment, and common friends on the social network.

A novel watermark technique for relational databases

In this paper, a new approach for protecting the ownership of relational database is presented. Such approach is applied for protecting both textual and numerical data. This is done by adding only one hidden record with a secret function. For each attribute, the value of this function depends on the data stored in all other records. Therefore, this technique is more powerful against any attacks or modifications such as deleting or updating cell values. Furthermore, the problems associated with the work in literature are solved. For example, there is no need for additional storage area as required when adding additional columns especially with large databases. In addition, in case of protecting data by adding columns, we need to add a number of columns equal to the number of data types to be protected. Here, only one record is sufficient to protect all types of data. Moreover, there is a possibility to use a different function for each field results in more robustness. Finally, the proposed technique does not have any other requirements or restrictions on either database design or database administration.

A Learning System for a Computational Science Related Topic

Abstract Computational science is an interdisciplinary field in which mathematical models combined with scientific computing methods are used to study systems of real-world problems. One of such mathematical models is automata theory. This paper introduces a learning system for automata theory. The learning system is a combination of java and robots technologies. Learners can build their own automaton graphically in the systems’ interface, and then pass it to the robot, which can then simulate the automaton transitions. Learners can learn by observing the robots motion. A preliminary evaluation shows the effectiveness of the system in classroom.

A Game-based Learning System for Theory of Computation Using Lego NXT Robot

Abstract Finite state automata are in the core of theory of computation course and related courses such as discrete mathematics, formal languages, etc. This paper introduces a finite state automata simulator and a robot-based game associated with it for active learning in theory of computation related courses. The simulator is implemented in Java language and the automaton game based robot is build by the Lego NXT Robot set. Learners can build their own automaton graphically in the simulator interface, and then pass it to the robot, which can then simulate the automaton transitions. Learners can learn by observing the robots motion.

Supporting Materials for Active e-Learning in Computational Models

In traditional lecture-driven learning, material to be learned is often transmitted to students by teachers. That is, learning is passive. In active learning, students are much more actively engaged in their own learning while educators take a more guiding role. This approach is thought to promote processing of skills and knowledge to a much deeper level than passive learning. In this paper, a research using supporting materials for active e-learning in computational models and related fields is presented. The contributions of this paper are supporting active tools to improve learning and an evaluation of its use in context.

A classroom experiment for teaching automata

In this work we focus on an experiment held at an automata class room to test the effectiveness of using simulators. We developed our own simulator. We also tested two other existing simulators to find whether simulators are useful as a teaching tool in automata classes.